Stylolites, Porosity, Depositional Texture, and Silicates in Chalk Facies Sediments

Home , Chalk

Downloaded from orbit.dtu.dk on: Dec 19, 2017

Stylolites, porosity, depositional texture, and silicates in chalk facies sediments

Ontong Java Plateau - Gorm and Tyra fields, North Sea

Fabricius, Ida Lykke; Borre, Mai K.

Published in: Sedimentology

Link to article, DOI: 10.1111/j.1365-3091.2006.00828.x

Publication date: 2007

Link back to DTU Orbit

Citation (APA): Fabricius, I. L., & Borre, M. K. (2007). Stylolites, porosity, depositional texture, and silicates in chalk facies sediments: Ontong Java Plateau - Gorm and Tyra fields, North Sea. Sedimentology, 54, 183-205. DOI: 10.1111/j.1365-3091.2006.00828.x

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Sedimentology (2007) 54, 183–205 doi: 10.1111/j.1365-3091.2006.00828.x

Stylolites, porosity, depositional texture, and silicates in chalk facies sediments. Ontong Java Plateau – Gorm and Tyra ﬁelds, North Sea

IDA L. FABRICIUS and MAI K. BORRE1 Institute of Environment and Resources, Technical University of Denmark, Bygningstorvet 115, DK-2800 Kgs. Lyngby, Denmark (E-mail: [email protected])

ABSTRACT Comparison of chalk on the Ontong Java Plateau and chalk in the Central North Sea indicates that, whereas pressure dissolution is controlled by effective burial stress, pore-filling cementation is controlled by temperature. Effective burial stress is caused by the weight of all overlying water and sediments as counteracted by the pressure in the pore fluid, so the regional overpressure in the Central North Sea is one reason why the two localities have different relationships between temperature and effective burial stress. In the chalk of the Ontong Java Plateau the onset of calcite-silicate pressure dissolution around 490 m below sea floor (bsf) corresponds to an interval of waning porosity-decline, and even the occurrence of proper stylolites from 830 m bsf is accompanied by only minor porosity reduction. Because opal is present, the pore-water is relatively rich in Si which through the formation of Ca–silica complexes causes an apparent super-saturation of Ca and retards cementation. The onset of massive pore-filling cementation at 1100 m bsf may be controlled by the temperature-dependent transition from opal-CT to quartz. In the stylolite-bearing chalk of two wells in the Gorm and Tyra fields, the nannofossil matrix shows recrystallization but only minor pore-filling cement, whereas microfossils are cemented. Cementation in Gorm and Tyra is thus partial and has apparently not been retarded by opal-controlled pore-water. A possible explanation is that, due to the relatively high temperature, silica has equilibrated to quartz before the onset of pressure dissolution and thus, in this case, dissolution and precipitation of calcite have no lag. This temperature versus effective burial stress induced difference in diagenetic history is of particular relevance when exploring for hydrocarbons in normally pressured chalk, while most experience has been accumulated in the over-pressured chalk of the central North Sea. Keywords Cementation, chalk, diagenesis, North Sea, Ontong Java Plateau, porosity, stylolites.

INTRODUCTION role (e.g. Scholle, 1977; Bathurst, 1995), but it remains an issue of debate how temperature and Cementation of chalk during burial diagenesis pressure controls this process, and whether clay requires a source of carbonate and it is generally minerals promote the process (e.g. Heald, 1955; accepted that pressure dissolution plays a major Simpson, 1985; Lind, 1993a; Sheldon et al., 2003; Safaricz & Davison, 2005). It also remains unclear 1Present address: Mærsk Olie og Gas A/S, Copen- how diagenetically controlled changes in the hagen, Denmark. mineralogy of silicates in the chalk inﬂuence

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists 183 184 I. L. Fabricius and M. K. Borre changes in the dominating carbonate phase SAMPLES AND METHODS (Baker et al., 1980; Hobert & Wetzel, 1989). Finally presence of hydrocarbons may have an Ontong Java Plateau influence on the diagenetic process (Scholle, 1977). Ooze and chalk with colour bands as well as Diagenetic processes are controlled by tem- chalk and limestone with stylolites were sampled perature, burial stress and pore-pressure, so, by from Site 807 during ODP Leg 130 (Fig. 1). Part of comparing two settings where some but not all each sample was impregnated with epoxy resin of these conditions differ, the understanding of and polished before backscatter electron micro- chalk diagenesis and porosity development graphy using a Philips XL40 scanning electron during burial should be improved. The two microscope (Philips, Holland). Silica and barite settings are the chalk facies sediments and were analysed by wavelength dispersive micro- sedimentary rocks of the Ontong Java probe analysis using a JEOL Superprobe 733 Plateau as found in Ocean Drilling Program (JEOL, Japan). Standards were: quartz, barite, (ODP) Leg 130 Site 807 (no hydrocarbons) and celestite, and calcite. chalk of the Central North Sea as found in two Another part of each sample was soaked in wells of the Gorm and Tyra fields (hydrocarbon demineralized water and slowly leached by acetic bearing). acid, while controlling the pH to be above 5Æ5in The main components of pelagic chalk are order to prevent dissolution of all non-carbonates. calcareous nannofossils and microfossils. Due to The filtrate was analysed for Al, Fe, Mg, Si, and Sr the contrast in size between these fossils, micro- by Atom Absorption Spectroscopy (Perkin-Elmer fossils reside in a continuous matrix of nanno- 5000, CT, USA). The residue was analysed for fossils. In the North Sea area, chalk facies mineralogical composition by X-ray diffraction by sediments of the Upper Cretaceous and Palaeo- use of Cu K-a radiation over an interval of 2Q 2 to gene Chalk Group are commonly referred to as 60. Diffractograms were also recorded from 2 to ‘chalk’ irrespective of degree of induration. This 30 subsequent to 3 days of glycolation at 60 C, may cause confusion when comparison is made and subsequent to heating to 300 C and to 500 C to pelagic deep sea carbonate sediments. These respectively in order to identify clay minerals. sediments and sedimentary rocks of Cretaceous Image analysis was done on backscatter elec- to Present age are of chalk facies and very tron micrographs of impregnated polished sam- similar in appearance and composition to North ples also studied by Borre & Fabricius (1998), to Sea chalk, but they are described as carbonate measure nannofossil matrix porosity, and ooze, chalk or limestone as a reflection of amounts of large particles (e.g. foraminifers) and increasing induration (Mazzullo et al., 1988). In pores by use of the filtering method described by the discussion of diagenesis a distinction is Borre (1997; Figs 2 and 3). Samples, visually free made between recrystallization and pore-filling of stylolites or colour bands, were analysed for cementation. Recrystallization involves local specific surface area using N2 adsorption (BET) simultaneous dissolution of carbonate particles with a MicromeriticsGemini III 2375 surface area and overgrowth on carbonate particles, where no analyser (GA, USA). 2+ 2 net transport of Ca and CO3 is required and no porosity change necessarily is involved, Gorm and Tyra fields whereas pore-filling cementation is a result of diffusion and net precipitation of material Plug samples from Gorm Field well N-22X and derived, e.g. from dissolution at stylolites (Borre Tyra Field well E-5X (Fig. 1) were chosen, so that & Fabricius, 1998). for most samples He-porosity data were available The present paper includes a discussion of the from the operator. Ultrasonic velocity data of the role of temperature and stress on stylolite forma- samples were discussed by Borre (1998). The tion, and the role of clay minerals in pressure majority of the Gorm Field samples were dis- dissolution. It also discusses the mainly porosity- cussed with respect to specific surface and per- neutral process of recrystallization, as well as meability by Mortensen et al. (1998). The Tyra pore-filling cementation, and how these processes samples were discussed with respect to image are related to depositional texture and the non- analysis by Røgen et al. (2001), and with respect carbonate components of the chalk. Finally the to permeability and capillary entry pressure by effect of these processes on porosity and perme- Røgen & Fabricius (2002). In the present study ability is discussed. backscatter electron micrographs from all se-

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 185

Fig. 1. Location of studied wells: The Ontong Java Plateau, Ocean Drilling Program Leg 130, Site 807, Gorm Field well N-22X, and Tyra Field well E-5X (modified after http://www-odp.tamu.edu/sitemap/ sitemap.html). lected samples were analysed to measure matrix large pores (Fig. 3; Bachman, 1984). Below porosity and amount of large particles and pores 1135 m below sea floor (bsf) in the limestone by use of the filtering method described by Borre interval, several beds of re-deposited sediment (1997; Fig. 3); and to measure specific surface by were recorded, as indicated by intra-clasts, silicate use of the filtering method described by Borre clasts, and slump structures (Kroenke et al., 1991). et al. (1997). The temperature of the OJP ranges from around 0 C at the sea bottom to around 20 C at 1 km bsf (estimate based on wireline logging data) and the CASE SETTINGS lower part of the OJP chalk is under an effective vertical stress of 8–10 MPa (Fabricius, 2003). ODP Site 807 is especially suited for the study Ontong Java Plateau – description of facies and of burial diagenesis due to the presumed simple case setting subsidence history and drainage equilibrium of The Ontong Java Plateau is a large oceanic basaltic the OJP (Table 1). When interpreting the dia- plateau situated in the western Pacific Ocean genetic history from well data in the Ontong Java (Fig. 1). The thick cover of pelagic carbonates Plateau, it is assumed that primary variations in holds no accumulations of hydrocarbons, but has nannofossil assemblages and size can be disre- been penetrated several times by the Deep Sea garded. Porosity-decline in the upper 600 m of Drilling Project (DSDP) and the ODP (Winterer ooze and chalk is caused by mechanical compac- et al., 1971; Moberly et al., 1986; Kroenke et al., tion, as confirmed by loading experiments (Fig. 4; 1991; Mahoney et al., 2001). ODP Site 807 cored Lind, 1993b; Fabricius, 2003). Simultaneously more than 1300 m of pelagic carbonate sediments with mechanical compaction, recrytallization of spanning the entire range from Late Cretaceous to calcite crystals causes particles to become more present. The upper 293 m of the sediment is equant, making way for further mechanical com- composed of nannofossil ooze, overlying 805 m of paction (Fig. 4; Borre & Fabricius, 1998). Image nannofossil chalk followed by 253 m of nanno- analysis of backscatter electron micrographs in- fossil limestone (Fig. 2; Kroenke et al., 1991). The dicates that the recrystallization causes the sur- main component of the sediments is calcareous face relative to volume (specific surface) of nannofossils, which constitute the matrix of the particles to decline; however, the compaction mud- and wackestones. The second most import- counteracts this effect on pores and causes spe- ant constituent is calcareous microfossils, which cific surface of pores to decline only a little. may be seen as composed of large particles and Below 600 m bsf follows an interval of near

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 186 I. L. Fabricius and M. K. Borre

Fig. 2. Ontong Java Plateau, Leg 130, Site 807. Recrystallization and pore-ﬁlling cementation. Scale bar is 16 microns. Backscatter electron micrographs: calcite is light grey and pore space is black. (For letters refer to Fig. 4.) (b) 807A-11H-5; Lower Pliocene, depth: 100 m bsf, CaCO3:93Æ1%, /lab:65Æ5%, /matrix:66Æ8%. Ooze, particles are in physical contact and they are partly recrystallized. (h) 807C-6R-1; Lower Oligocene, depth: 829 m bsf, CaCO3:94Æ9%, /lab:55Æ1%, /matrix:58Æ3%. Chalk, particles are connected by internal cement resulting from recrystallization, which has also allowed particles and pore size to increase. (n) 807C-48R-1; Upper Palaeocene, depth: 1146 m bsf, CaCO3: 97Æ9%, /lab:14Æ8%, /matrix:19Æ8%. Limestone, microfossils and matrix pore space partly ﬁlled by cement, probably externally derived from stylolites.

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 187

ponents, the upper 490 m of the ooze and chalk contain up to centimetre-wide green and purple colour bands (Kroenke et al., 1991), the colouring of green bands is due to finely disseminated green silicates probably of volcanic origin, the colouring of purple bands is due to disseminated pyrite (Lind et al., 1993). The distribution of bands may Fig. 3. Conceptual sketch of backscatter electron reflect variations in plankton productivity (Berger micrographs of chalk (Røgen, 2002). Left: At high & Lind, 1997). Below 490 m bsf the colour bands magnification small particles (white) and small pores are replaced by more distinct coloured flaser (black) are seen. They constitute the chalk matrix, and porosity as measured in this image will be the matrix structures, these are again replaced by grey flaser porosity. Right: At low magnification the matrix is structures and stylolites below 830 m bsf (Figs 5 blurred, whereas large particles and cemented areas are and 6; Lind, 1993a). The occurrence of stylolites white, and large porosity is black. Upon repeated fil- is associated with the onset of pore filling calcite tering, the amount of large particles and cemented cementation in equilibrium with pore water as areas, as well as large pores can be measured by image reflected in the trend to lower d18O values in bulk analysis (Borre, 1997). carbonate samples (Fig. 6; Lind, 1993a). Opaline chert is found from 959 m bsf in the lower part of the chalk section and in the limestone (Fig. 6; Table 1. Diagenetic processes in chalk facies sedi- Kroenke et al., 1991). ments and sedimentary rocks of Ontong Java Plateau. Depth (m bsf) North Sea chalk, Gorm and Tyra fields – 0–293 Pelagic carbonate ooze lithology description of facies and case setting 0–600 Mechanical compaction causes Gorm and Tyra fields are structural traps in the porosity decrease chalk of Danish North Sea. Gorm is an oil field 0–490 Sediment contains green and purple colour bands with Maastrichtian chalk as the main reservoir 0toca 200 Recrystallization dominates specific (Hurst, 1983), whereas Tyra is primarily a gas surface of particles and pores, below field with Danian chalk as the main reservoir this depth decreasing influence along (Doyle & Conlin, 1990). with that of mechanical compaction Gorm field well N-22X cored 122 m of chalk of ca 200–600 Mechanical compaction dominates Late Cretaceous (Maastrichtian) and Palaeogene specific surface of particles and pores (Danian) age, whereas Tyra field well E-5X cored 293–1098 Chalk lithology 490–830 Sediment contains coloured flaser 24 m of chalk of Palaeogene (Danian) age. The structures chalk of the North Sea resembles the OJP chalk by 600–950 Porosity roughly constant, while pores containing well-sorted mudstones as well as less and particles grow well-sorted wackestones. The Gorm and Tyra 780–950 Transition from opal-A to opal-CT fields are structures with relatively low relief, so 830– Sediment contains grey flaser that a heavy influence on diagenesis from advect- structures and stylolites ing waters seen over salt piercement structures is 959– Sediment contains chert 1098– Limestone lithology not expected (Jensenius, 1987). Please note that 1127– Quartz is the only silica phase the chalk fields of the North Sea vary with respect to degree of recrystallization and pore-filling cementation, so the two wells in Gorm and Tyra fields are not necessarily typical. constant porosity, where declining specific sur- Due to the regional overpressure in the central face of pores and particles indicates increasing North Sea (e.g. Japsen, 1998), the effective stress particle- and pore size resulting from recrystalli- in the North Sea chalk of the Gorm and Tyra fields zation (Fig. 4; Fabricius, 2003). Around 1100 m is comparable with the effective stress at the bsf, pore-filling cementation by continuous over- deeper interval of the OJP (ca 10 MPa). They growth causes porosity to decline and specific differ though with respect to temperature and age. surface of particles to fall, while specific surface Mainly due to the deeper burial of the North Sea of pores rises (Fig. 4; Borre & Fabricius, 1998). chalk, the temperature is higher (ca 70 C) than at The carbonate of Ontong Java Plateau is low-Mg OJP. Loading experiments on samples from calcite. With respect to the non-carbonate com- nearby Gorm field well indicate that this chalk

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 188 I. L. Fabricius and M. K. Borre

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 189

Fig. 4. Data from ODP Site 807. Porosity is from shipboard physical properties data on discrete samples (dots) and calculated from the density log (solid line) (Kroenke et al., 1991). Composition is from analysis of BSE-images. Signatures from left to right indicate; light grey: silicates, white: large particles or large cemented area, hatched: small particles, dark gray: large porosity, medium grey: small porosity (Fig. 2 and 3). Matrix porosity is from image analysis (Figs 2 and 3; Borre & Fabricius, 1998), intraclast-bearing samples are indicated by small bars. BET is specific surface as measured by N2 adsorption. Spar and S/ are specific surface as measured by image analysis with reference to particles, respectively porosity (Borre & Fabricius, 1998). does not undergo mechanical compaction at its genic barite (Figs 9 and 11). Dolomite was not present effective burial (Olsen, 2002). detected. The chalk of the Gorm and Tyra fields is From around 830 m bsf the flaser structures are composed of low-Mg calcite. It contains stylolites grey. They are now easily visible in backscatter resembling those of the OJP with respect to drape electron micrographs (Fig. 8). Well-developed thickness and amplitude (Fig. 7). In the Gorm stylolites with columns and clay drapes are found Field stylolite drapes are primarily composed of from 830 m bsf (Fig. 8). The dominating non- kaolinite and quartz (Lind et al., 1994), but non- carbonates are still smectite and opal, but from carbonates of the Tyra and Gorm fields include around 780 m bsf down to around 950 m bsf, smectite as well as kaolinite and quartz (Røgen & opal-A is gradually replaced by opal-CT. Fabricius, 2002). Silica and silicates RESULTS In the entire carbonate section cored in ODP Site 807, quartz is found as a minor, probably wind- borne constituent (Krissek & Janecek, 1993), but Colour bands, flasers and stylolites of Ontong biogenic opal-A dominates in the upper section, Java Plateau before being replaced by opal-CT in the depth The green colour bands of the ooze and upper interval 780–950 m bsf. In this depth interval, the chalk section of the OJP are difficult to recognize chalk has an Oligocene age corresponding to 31– on backscatter electron micrographs (Fig. 8). 43 Ma (Kroenke et al., 1991), and an estimated Nevertheless, detailed microprobe analysis al- temperature of 15–20 C. In line with the obser- lows the identification of dispersed, up to vations of Hobert & Wetzel (1989), the transition 10 micron-size Fe-bearing, possibly volcanigenic from opal-A to opal-CT is accompanied by a fall clay particles (Lind et al., 1993). The following in specific surface of the sediment as measured by minerals were identified in colour band-samples BET (Fig. 4). by X-ray diffraction: smectite-chlorite (all sam- Down to around 950 m bsf the amount of ples) which probably gives rise to the green dissolvable Si decreases in the sediment, con- colouring; quartz, albite (most samples) and illite current with an increase in Si-concentration in (single samples), which probably are air-blown the pore water (Fig. 9). The insignificant role of (Krissek & Janecek, 1993); as well as barite and silicates for the Si concentrations is indicated by pyrite (single samples) with crystal shape indica- a low Al content (generally below 0Æ001%) in the ting authigenic origin. All samples contain bio- dissolvable sediment. No analysis for Al was genic opal-A (Figs 6 and 8–10). included in the shipboard pore water analyses With depth, the colour bands become more (Kroenke et al., 1991). Opal-A is thus progres- distinct and wispy until from 490 to 830 m bsf sively going into dissolution with depth. From they may be described as green-coloured flaser microprobe analysis, biogenic opal-A was found structures (Fig. 5). Concurrently, the chlorite- to have 85–86% SiO2 at 894 and 919 m bsf peak disappears from the X-ray diffractograms (Table 2; Fig. 10). Below 900 m bsf the concen- and the clay seems to be only smectite. These tration of silica in the pore water decreases, and flaser structures are discernible in backscatter the first chert nodules were noticed at 959 m bsf electron micrographs (Fig. 8). Dissolution of cal- (Fig. 6; Kroenke et al., 1991), indicating precipi- citic microfossils at contacts with silicates is tation of opal-CT in line with e.g. Kastner (1979). visible. The non-carbonate fraction of the sedi- At 1007 m bsf, Opal-CT was found as fossil- ment is dominated by smectite and biogenic filling cement and as replacement of radiolarian opal-A; and in addition small amounts of prob- remains in both cases with 94% SiO2 (Table 2, ably airborne quartz and albite, as well as authi- Fig. 10). At 1098 m bsf massive silica with 98%

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 190 I. L. Fabricius and M. K. Borre

Fig. 5. Development of stylolites in Ontong Java Plateau chalk. Scale is in centimeters. (For letters refer to Fig. 4.) (a) Diffuse colour bands, 0.5 cm thick green band (darkest) with purple band (lighter) of similar thickness immediately below and a thinner purple band 0.5 cm above, 807B-5H-5; Upper Pliocene, depth: 38 m bsf, /lab:67Æ1%. (f) Green- coloured ﬂaser, 807A-75X-2; Lower Oligocene, depth: 710 m bsf, /lab:50Æ9%. (g) Grey ﬂaser, 807C-4R-2; Lower Oligocene, depth: 811 m bsf, /lab:48Æ5%, (o) Stylolite, 807C-56R-1; Upper Maastrichtian, depth: 1207 m bsf, /lab: 13Æ1%.

)1 SiO2 was found (Table 2, Fig. 10). Below 1127 m 22 mmol l at 350 m bsf and remains around bsf quartz is the only observed silica-phase, and that concentration below that depth (Kroenke little dissolvable Si is left in the sediment et al., 1991). The decline is probably partly due to (Fig. 9). sulphate-reducing bacteria giving rise to the Below the depth of onset of pore filling calcite formation of pyrite. Fe for the pyrite is partly cementation the following changes are seen in derived from bacterially reduced magnetite (Mus- silicate mineralogy. In the interval 958–1181 m grave et al., 1993) and partly derived from the bsf clinoptilolite is found. The first occurrence of clay particles of the green colour bands, which the zeolite clinoptilolite corresponds to the depth may turn purple due to colouring by dissemin- where opal-A has disappeared. From 1181 m bsf ated pyrite (Lind et al., 1993). The content of (20–25 C) smectite-illite becomes the dominating dissolvable Fe in the sediment was found to be clay mineral, and below that depth the feldspar below 0Æ001%. albite is replaced by anorthite. Another sink for sulphate is barite, which was noted from 191 m bsf. The barite was found as 1– 10 micron size probably authigenic crystals Sulphates and sulphides (Fig. 11). Microprobe analysis indicates an Sr/Ba Sulphate concentration in the pore water declines atomic ratio of around 0Æ03 in the barite crystals ) from ca 27 mmol l 1 near the sea floor to around (Table 2). Sr in the pore water increases until a

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 191

Fig. 6. Occurrence of green and purple colour bands, flaser structures, and stylolites in ODP Site 807, when compared with carbonate content in sediment (left) and d18O (right) (this study; Kroenke et al., 1991; Lind, 1993a). depth of around 200 m bsf, below which depth it ated lower part of the limestone (Fig. 4). gradually decreases (Fig. 9). Intraclast-bearing limestone has porosity comparable with limestone with similar microtexture and no visible clasts (Fig. 4). The non-carbonate Texture and effect of cementation on Ontong fraction is mainly diagenetic quartz which acts as Java Plateau and North Sea fields a part of the pore-filling cement. Petrographic data from the OJP show how pore The chalk samples of Gorm and Tyra wells are filling cement primarily fills the pores of micro- less porous and more recrystallized than the fossils (Fig. 2). Image analysis data show how the Neogene and Palaeogene OJP chalk, as reflected pore-filling cementation affects porosity (Fig. 4). in the lower specific surface of particles (Spar)in Above the depth of onset of widespread pore- North Sea chalk than in OJP chalk (Figs 2, 4 and filling cementation around 1100 m bsf, large 12–15). The chalk of the Gorm and Tyra wells is intrafossil pores and large microfossil shells of early Palaeogene (Danian) and Cretaceous contribute roughly equally to the chalk volume, (Maastrichtian) age and thus contemporary with so changes in bulk porosity mainly reflect chan- the limestone of the OJP. The calcite crystals ges in matrix porosity (Fig. 4). Variations in chalk building up the chalk of Maastrichtian age in both texture as defined from variations in microfossil places appear to be recrystallized to the same content thus only have little influence on bulk extent, as reflected in euhedral outline and low porosity in the sediments in the interval without specific surface of particles (Spar) as measured by major pore-filling cementation. The non-carbon- image analysis (Figs 2, 4, 12 and 14). The North ate fraction in the ooze and chalk intervals is Sea chalk of Danian age has higher specific mainly biogenic opal. surface and seems less recrystallized than the Below 1100 m bsf pore-filling cementation is OJP limestone (Figs 2, 4 and 12–15). The chalk of more pronounced in wackestones than in mud- Gorm and Tyra wells is significantly more porous stones, so that the wackestone-dominated upper (porosity: 20–40%) than the limestone of the OJP part of the pelagic limestone interval has mark- (porosity: 10–25%), as also reflected in a smaller edly lower porosity than the mudstone-domin- specific surface of pores (S/) in the North Sea

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 192 I. L. Fabricius and M. K. Borre

Fig. 7. Stylolites from Gorm Field. As seen in core (D1 and E1 – scale is in centimeters) and as seen in backscatter electron micrographs (D2 and E2 – scale bar is 100 microns); pyrite is white, calcite is light grey, magnesium-bearing calcite is darker grey, silicates are dark gray, and pore space is black. The near horizontal fractures are caused by sampling (for letters refer to Fig. 14). Both samples are from Upper Maastrichtian chalk in well N-22X. Sample D is from a depth of 2167 m sub sea and has a porosity of 24Æ6%, sample E is from a depth of 2169 m sub sea and has a porosity of 23Æ5%. The clay drape in both samples are dominated by kaolinite and quartz with minor amounts of pyrite, plagioclase, and barite (authigenic as well as from drilling mud) (Lind et al., 1994). chalk when compared with the OJP limestone DISCUSSION (Figs 4, 14 and 15). In the samples of the Gorm field, pore-filling Stylolite formation – role of temperature and cementation of microfossils is almost complete in stress chalk of Maastrichtian age, whereas in samples from Gorm and Tyra fields, cementation of Pressure dissolution at stylolites is a major source microfossils is only partial in chalk of Danian of carbonate for pore-filling cementation (e.g. age (Figs 12–15). The matrix of the Gorm Field Sheppard, 2002), nevertheless, it is not generally chalk samples is recrystallized as indicated by agreed whether the pressure dissolution of chalk euhedral outline of the grains, but only weakly is mainly controlled by stress at the grain contacts affected by pore-filling cementation as indicated or by temperature dependent diffusion and pre- by a matrix porosity in Maastrichtian chalk of the cipitation (Renard et al., 2000). Bjørkum (1996) Gorm Field being close to the matrix porosity of and Bjørkum et al. (1998) interpreted petro- OJP chalk (Figs 4 and 14). In the Danian chalk graphic data as indicating a temperature control samples of the Gorm and Tyra wells, pore-filling on stylolite formation in quartzose sandstones, submicron-size quartz and clay reduce matrix and supported their conclusion by referring to porosity (Figs 12–15). geochemical modeling data by Oelkers et al.

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 193

Fig. 8. Development of stylolites in Ontong Java Plateau chalk. Scale bar is 16 microns. Backscatter electron micrographs: calcite is light grey, silicates are darker grey and pore space is black. (For letters refer to Fig. 4.) (c) Diffuse colour band, 807A-12H-3; Lower Pliocene, depth: 107 m bsf, CaCO3:92Æ2%, /lab:64Æ4% /matrix:64Æ1%, non- carbonates: opal-A, chlorite-smectite, quartz, albite. (e) Coloured ﬂaser, 807A-74X-2; Lower Oligocene, depth: 699 m bsf, CaCO3:94Æ4%, /lab:50Æ6%, /matrix:52Æ7%, non-carbonates: opal-A, smectite, quartz, albite. The near horizontal fracture is caused by sampling. (g) Grey ﬂaser, 807C-4R-2; Lower Oligocene, depth: 811 m bsf, CaCO3:90Æ5, /lab: 48Æ5%, /matrix:50Æ1%, non-carbonates: opal-CT, smectite with minor chlorite, quartz, barite, albite. (o) Stylolite, 807C-71R-2; Upper Campanian-Lower Maastrichtian, depth: 1350 m bsf, CaCO3 >62Æ8%, /lab:23Æ6%, /matrix:19Æ2%, non-carbonates: quartz, smectite-illite, anorthite. Irregular light clasts are apatitic skeletal remains.

(1992) indicating a strong temperature-depend- 1350 m bsf, the temperature is around 25 C. The ence of quartz dissolution and precipitation rate chalk of the Gorm and Tyra fields is at present constants as well as aqueous diffusion coeffi- much warmer (ca 70 C), and the isotopic com- cients. Dissolution of calcite is not promoted by position of the chalk of the Tyra Field [d18Oof)4 increasing temperature so a similar temperature to )2 (permil) PDB; Røgen & Fabricius, 2004] dependence on stylolite formation in chalk is not indicates that recrystallization has taken place at to be expected. a higher temperature than in the OJP, where d18O The issue can be addressed by noting that the generally is higher (less negative) than )2 (per- stylolites of the OJP resemble those of the Gorm mil) (Fig. 6). These data thus indicate a control by and Tyra fields with respect to amplitude (Figs 7 stress rather than by temperature for stylolite and 8). The effective stress of the North Sea chalk formation in chalk. and the stylolite-bearing chalk of the North Sea fields is similar (around 10 MPa), whereas the Stylolite formation – role of sheet silicates temperature differs. In the OJP, at 830 m bsf where stylolites start forming, the temperature is Whereas several authors have envisaged that around 20 C, at the base of the limestone at stylolites develop from pressure dissolution of

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 194 I. L. Fabricius and M. K. Borre

Fig. 9. ODP Site 807. Above: Concentrations of Si, Mg, and Sr in acetic acid-dissolvable part of the sediment. Solid symbols are results of analysis, open symbols are sediment composition corrected for material dissolved in pore water. The insoluble residue was analysed by X-ray diffraction, and occurrence of minerals was noted. In addition to the shown minerals most samples contain quartz. Below: concentration of Si, Mg, Ca, and Sr in pore water (Delaney et al., 1991). Excess Ca in pore water was calculated by PHREEQC relative to equilibrium for calcite (Parkhurst & Appelo, 1999).

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 195

Fig. 10. Ontong Java Plateau, Leg 130, Site 807. Transition from opal to quartz. Scale bar is 10 microns. Backscatter electron micrographs: barite is white, calcite is light grey, silica is darker grey and pore space is black. (For letters refer to Fig. 4 and Table 2.) (i) Lower Oligocene, depth: 894 m bsf. Silicious fossil contains 86% SiO2. (j) Upper Eocene, depth: 919 m bsf. Silicious fossil contains 85% SiO2. (k) Middle Eocene, depth: 1007 m bsf. Silica contains 94% SiO2. (l) Middle Eocene, depth: 1098 m bsf. Silica contains 98% SiO2.

but the clay drape on the stylolite may indicate that the fracture has been coated with phyllosilicates prior to stylolite formation. Safaricz & Davison (2005) studied the volumetric effect of pressure dissolution at two localities in the North Sea chalk. By assuming that the incipient stylolites were randomly distributed in chalk, they calculated a chalk sediment column short- ening of around 50% and that around half the dissolved calcite was reprecipitated as cement, whereas the other half should have left the system. This would require a fluid flux so Fig. 11. Ontong Java Plateau, Leg 130, Site 807. enormous (a minimum of 8000 unit volumes Strontium-bearing barite. Scale bar is 10 microns. of fresh water per unit volume of preserved Backscatter electron micrographs: barite is white, cal- chalk in the Machar field; Safaricz & Davison, cite is light grey, silica is darker grey, and pore space is 2005), that it must be considered unrealistic. black. (For letters refer to Fig. 4 and Table 2.) (d) Upper The conclusion may be that the basic assump- Miocene, depth: 228 m bsf. Barite crystal has Sr/Ba of tion of a random stylolite distribution is not 0Æ055. (m) Lower Eocene, depth: 1127 m bsf. Barite valid. crystals have Sr/Ba of 0Æ031. Baker et al. (1980) showed that in pressure dissolution experiments with calcite, the pres- calcite–calcite contacts (or quartz–quartz in ence of clay prevents recrystallization of calcite. sandstones) (e.g. Scholle, 1977; Bathurst, 1995; Their data indicate that the clay inhibits the Sheldon et al., 2003), petrographic evidence precipitation of calcite, but also that clay does not frequently points to a role of sheet-silicates. retard the dissolution of calcite. This indicates Heald (1955) and several later authors that the contact between calcite and clay is a including Bjørkum (1996) found stylolites in place of disequilibrium for calcite where disso- sandstone to be localized at quartz–phyllosili- lution but not precipitation is possible. cate contacts. Bathurst (1995) concluded that Weyl (1959) advanced the idea that pressure localization at phyllosilicates is not necessary dissolution takes place via a thin water film in for stylolite formation, and sketched three the contact between crystals. The dissolved ions field observations showing that stylolites do should then be transported to the pores in the not always follow bedding. Similar obser- water film. This mechanism is simpler at a vations were done by Simpson (1985). In OJP calcite–clay interface than at a calcite–calcite stylolites may follow fractures or biogenic struc- interface, according to the following argument: tures at an angle to bedding (Lind, 1993a), recrystallization creates a smooth interface

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 196 I. L. Fabricius and M. K. Borre

Table 2. Wavelength dispersive microprobe analysis data of single mineral grains.

Depth Sr/Ba (atomic Sr/Ba (standard No. of analysed Sample (m bsf) ratio) (average) deviation) particles Barite 807A, 25H-2, 79–81 cm (d, Figs 4 and 11) 228 0Æ055 0Æ052 3 807A, 62X-5, 7–9 cm 588 0Æ068 0Æ067 4 807C, 2R-2, 40–42 cm 792 0Æ033 0Æ018 3 807C, 31R-1, 17–19 cm 1007 0Æ034 0Æ013 4 807C, 44R-1, 30–32 cm 1116 0Æ030 0Æ003 3 807C, 45R-2, 3–5 cm (m, Figs 4 and 11) 1127 0Æ031 0Æ005 3 807C, 46R-1, 40–42 cm 1136 0Æ038 0Æ009 4 807C, 50R-1, 40–42 cm 1156 0Æ029 0Æ004 3

SiO2 (wt%) SiO2 (wt%) Silica (average) (standard deviation) 807C, 15R-1, 33–35 cm (i, Figs 4 and 10) 894 86Æ00Æ64 807C, 20R-1, 16–18 cm (j, Figs 4 and 10) 919 85Æ40Æ74 807C, 31R-1, 17–19 cm (k, Figs 4 and 10) 1007 94Æ21Æ510 807C, 41R-1, 47–49 cm (l, Figs 4 and 10) 1098 98Æ00Æ412 where two calcite grains are held together by a Fabricius, 1998) and influence the transport of strong monolayer water-bridge. Across the ions dissolved at stylolites. bridge, Ca and carbonate ions may be transpor- During recrystallization, calcite releases Sr ted from the calcite particle in the less stable (Baker et al., 1982; Delaney & Linn, 1993) and crystallographic orientation to the particle in the approaches stable crystal shape, a process which more stable orientation. This process does not is largely over for sediments of Palaeogene age seem to involve net dissolution of calcite or (Figs 4, 6 and 9). The released Sr precipitates as porosity loss under normal burial diagenetic a component in barite (Fig. 11) and because it conditions (Fabricius, 2003) whereas, at high gets trapped in insoluble barite a general depth- stress, pressure dissolution among calcite parti- wise fall in content of dissolvable Sr in the cles is well documented (Baker et al., 1980). sediment is seen (Fig. 9). Delaney & Linn (1993) Under normal burial diagenetic conditions, con- suggested that the formation of celestite could tact between calcite and clay under stress pro- explain the concurrent decline in sulphate and motes dissolution because it involves forcing the stabilization of Sr in the pore water, but no calcite into the ion exchange layer of the clay. Ca celestite was observed in the present study, so at ions can thus be transported from the solid least the main part of the dissolved Sr is taken interface to the solution via an ion exchange up by barite. Cronan (1974) cited authors conveyer belt. This approach is largely similar to reporting up to 3% Sr in solid solution in the conclusion of Renard & Ortoleva (1997), who microcrystalline barite in deep-sea sediments. pointed to thick fluid layers at sheet silicates, A diagenetic origin of barite at this pelagic site is giving rise to easier diffusion. Stylolites thus in accordance with the observations of Von tend to be localized at clay, because the cation Breymann et al. (1989) and Dymond et al. exchange capacity of the clay furthers dissolu- (1992). Ba is probably released during decay of tion and counteracts precipitation at the inter- organic matter from siliceous fossils (Bishop, face. 1988), in the present case taking place during sulphate reduction. A hydrothermal source of Ba and Sr is hardly likely in the geological setting Recrystallization on the OJP. The mineralogical constituents of the OJP sedi- The siliceous fossils (radiolarians and dia- ments and sedimentary rocks equilibrate with toms) originate as opal-A, but are gradually pore water as a function of temperature and time, dissolved, as indicated by etched outline and equilibration of one mineral influences the (Fig. 10i,j), and reprecipitated as opal-CT which other. These processes are not directly influen- develops into quartz (Figs 9 and 10k,l). Based cing porosity, but may facilitate mechanical on a compilation of data from 37 DSDP volumes compaction by smoothing particles (Borre & of rates and temperature of formation of opal-

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 197

Fig. 12. North Sea, Gorm Field, Well N-22X. Scale bar is 16 microns. Texture and carbonate content control porosity and degree of recrystallization. The ﬁne-grained non-carbonates include quartz and kaolinite. Backscatter electron micrographs: calcite is light grey, silicates are darker grey and pore space is black. (For letters refer to Fig. 14.) (A) N- 22X; Danian, depth: 2093Æ1 m sub sea, CaCO3:84Æ5%, /lab:30Æ2%, /matrix:45Æ9%. The silicate rich wackestone has high porosity and relatively low recrystallization of matrix particles. Silicates generally occur as submicron size particles. (B) N-22X; Danian, depth: 2114Æ0 m sub sea, CaCO3:89Æ6%, /lab:23Æ1%, /matrix:39Æ2%. The poorly sorted chalk packstone has low porosity. Microfossils tend to be cemented and silicates generally occur as submicron size particles. (C) N-22X; Maastrichtian, depth: 2127Æ1 m sub sea, CaCO3:98Æ2%, /lab:36Æ0%, /matrix:49Æ5%. The well- sorted chalk mudstone has high porosity. A high degree of recrystallization is indicated by the equant calcite crystals of the chalk matrix (C1).

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 198 I. L. Fabricius and M. K. Borre

Fig. 13. North Sea, Tyra Field, Well E-5X. Scale bar is 16 microns. Porosity is controlled by content of microfossils and content of carbonate. The non-carbonates include quartz and smectite. Backscatter electron micrographs: calcite is light grey, silicates are darker grey and pore space is black. (For letters refer to Fig. 15.) (a) E-5X; Danian, depth: 2019Æ9 m sub sea, CaCO3:94Æ5%, /lab:33Æ6%, /matrix:36Æ4%. The relatively pure and well-sorted chalk has high porosity. (b) E-5X; Danian, depth: 2027Æ0 m sub sea, CaCO3:86Æ7%, /lab:21Æ8%, /matrix:29Æ2%. The relatively impure and poorly sorted chalk has low porosity. A stylolite transects the image from lower left to upper right. The drape is composed mainly of smectite and quartz.

CT, the following relationship was found by lower geothermal gradients and/or less carbon- Kastner (1979): ate-rich sediments than Eq. 2. According to Eq. 2, 31–43 Ma corresponds to 13–21 C in accordance t ¼ 80 2T þ 001T2 ð1Þ with the present data. The transformation from opal-CT to quartz is accompanied by a drop in Si t ¼ 83 415T þ 01T2 0001T3 ð2Þ concentration in the pore water, which may also be inﬂuenced by precipitation of clinoptilolite- where t is time after deposition in millions of heulandite (Fig. 9). years and T is present day temperature (C). Eq. 1 The mineralogical recrystallization will also to represents slower sedimentation rates and/or some extent modify sedimentary structures.

Fig. 14. Data from well N-22X in Gorm Field. Porosity is from He-porosimetry on discrete samples (dots) and calculated from the density log (solid line). Composition is from analysis of BSE-images. Signatures from left to right indicate: light grey: silicates, white: large particles or large cemented area, hatched: small particles, dark gray: large porosity, medium grey: small porosity (Figs 3 and 12). Matrix porosity is from image analysis (Figs 3 and 12). BET is speciﬁc surface as measured by N2 adsorption. Spar and S/ are speciﬁc surface as measured by image analysis with reference to particles, respectively with reference to porosity.

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 199

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 200 I. L. Fabricius and M. K. Borre

Fig. 15. Data from well E-5X in Tyra Field. Porosity is from He-porosimetry on discrete samples (dots) and calculated from the density log (solid line). Composition is from analysis of BSE-images. Signatures from left to right indicate; light grey: silicates, white: large particles or large cemented area, hatched: small particles, dark gray: large porosity, medium grey: small porosity (Figs 3 and 13). Matrix porosity is from image analysis (Figs 3 and 13). BET is speciﬁc surface as measured by N2 adsorption. Spar and S/ are speciﬁc surface as measured by image analysis with reference to particles, respectively with reference to porosity.

Green colour bands contain smectite-chlorite, Calcite cementation – role of silica whereas coloured flasers contain smectite (Figs 6 and 9). In order to form the wispy structure, the Changes in silica-modification seem to influence clay minerals must have recrystallized, while calcite cementation. A control of calcite cemen- releasing Fe. Mg in pore water decreases with tation by silica was also reported by Hobert & depth at a declining rate, and thus probably not Wetzel (1989). They studied deep-sea carbonates only reflects diffusion of Mg from sea water to and found ‘carbonate sediments containing chert basement but also reflects uptake of Mg during (1) tend to be more indurated and display more this recrystallization, although a corresponding advanced diagenetic alteration, regardless of sub- increase in sediment Mg-content is less obvious bottom depth’. Correspondingly, Baker et al. (Fig. 9). The transition from opal-A to opal-CT (1980) found that diatoms inhibit calcite recrys- accompanies the development of grey flaser tallization. structures and stylolites as well as initial pore- The opal-A of diatoms and radiolarians will filling calcite-cementation, and the transition cause a relatively high concentration of Si in the from opal-CT to quartz accompanies the marked pore water, which will counteract carbonate increase in pore-filling calcite-cementation below cement precipitation for the following reason: 1100 m bsf. Tanaka & Takahashi (2000, 2002) found that

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 201 whereas all alkali and earth-alkali metal ions their relatively high porosity, the mudstones facilitate dissolution of silica, Sr, Ca, and Na ions cannot have formed by a gradual depth-wise form stable complexes with silicate ions in disappearance of foraminifers due to preferential chloride solutions comparable to sea water. Ca dissolution and precipitation, as suggested by and especially Sr ions form more stable com- Schlanger & Douglas (1974), but must rather be plexes than Na ions, so although the pore water of seen as having primary texture. ) ODP Site 807 contains 0Æ5 mol l 1 Na (Delaney Petrographic image analysis indicates that the et al., 1991), the concentration of Ca is high local porosity variations in the Gorm and Tyra ) enough (>10 mmol l 1, Fig. 9) for Ca–silicate fields are also due to differences in primary complexes to dominate. Sr concentrations are sediment composition. Low porosity may be too low to be significant in this context (Fig. 9). caused by a high content of cemented micro- Fabricius (2003) found that the pore water of ODP fossils or by a high content of pore-filling Site 807 is supersaturated with respect to calcite silicates. A combination of the two effects causes and explained it by adsorbed Mg-ions acting as a particularly low porosity in the lower part of the shield of Mg-calcite according to Neugebauer Danian section as seen in the data of the Tyra (1974) and Berner (1975). No Mg-calcite was Field (Figs 13, 15). observed, so in this case a more simple explanation for the apparent Ca-supersaturation is the Comparison of porosity and permeability, formation of Ca–silicate complexes in the pore Ontong Java Plateau and North Sea fields water. The Si-concentration in the pore water is indeed 2–4 times that of the excess Ca (Fig. 9), The porosity of the chalk of Gorm and Tyra fields allowing all excess Ca to be part of silica has a different textural distribution compared complexes. When opal-A is the dominating sil- with OJP chalk and limestone. Gorm and Tyra ica-phase, the pore water will be so rich in Si that chalk is similar to OJP limestone by having a part of the Ca released during recrystallization cemented microfossils, but differs in having and Ca generated by pressure dissolution will not larger matrix porosity and less pore-filling cemen- precipitate, but rather form Si-complexes in the tation of matrix (Figs 4, 14 and 15). Gorm and pore water or take part in the overall diffusion of Tyra chalk is similar to OJP chalk with respect to Ca from basement to sea water. No net calcite matrix porosity but differs by having cemented cementation was seen before silica in the pore microfossils (Figs 4, 14 and 15). water starts falling as a consequence of opal-CT The similar matrix porosity may be a reflection formation and a conspicuous calcium cementa- of the similar degree of (now arrested) mechanical tion front is seen when opal-CT has been replaced compaction. The mechanical compaction of OJP by quartz. Calcite cementation is thus probably seems to have arrested around 600 m bsf, below controlled by the diagenetic development of silica which depth sediment stiffening and increase in phases and thus temperature, and the frequently particle and pore size take place by internal reported apparent supersaturation of Ca in pore redistribution of calcite (recrystallization), before water in carbonate sediments may be a conse- the onset of pore-filling cementation (Fabricius, quence of not taking Ca–silica complexes into 2003). A similar arrest in mechanical compaction account in geochemical calculations. and subsequent pore-growth may have taken place in the North Sea fields before their present effective burial as indicated by a relatively high Calcite cementation and porosity – role of sonic P-wave velocity (Borre, 1998). depositional texture In spite of the relatively large pores and large Down to the cementation front in the OJP, porosity of the OJP, the permeability is probably variations in primary texture have little influence lower than in the North Sea chalk (Fig. 16). The on total porosity, but in the deep cemented reason is the relatively high specific surface as interval the influence of primary texture on measured by BET (Figs 4, 14 and 15). The porosity is significant (Fig. 4). Because micro- difference in BET is caused by the content of fossils are preferentially filled with cement, opal in the OJP chalk, whereas quartz is the only wackestones get lower porosity than mudstones. reported silica phase in the studied Gorm and This is the explanation for the reversing porosity Tyra chalk (Lind et al., 1994; Røgen & Fabricius, trend in the deepest part of ODP Site 807 (Fig. 4). 2002). The microfossil-rich interval thus has lower The lack of pore-filling inter-particle cementa- porosity than the underlying mudstones. Due to tion in the matrix of the chalk of the Gorm and

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 202 I. L. Fabricius and M. K. Borre

Fig. 16. Permeability of studied samples. Measured gas permeability is open circles. Black symbols represent permeability calculated from Kozeny’s equation and a relationship between gas permeability and liquid permeability as given in Mortensen et al. (1998).

Tyra fields indicates a more gradual cementation have taken place in the studied North Sea chalk when compared with the OJP, where a cementa- wells. tion front is seen. The differences may be In the cold Ontong Java Plateau net dissolution explained in several ways: of calcite is controlled by stress whereas net 1 The introduction of hydrocarbons into the precipitation is controlled by temperature. In the two North Sea structures may have impeded pore- warmer chalk of the central North Sea dissolu- filling cementation at a critical initial stage of tion as well as precipitation may be stress cementation, only allowing volume-preserving controlled because of relatively early transfor- recrystallization of matrix crystals, but, however, mation from opal to quartz. When exploring for allowing cement precipitation within the water- hydrocarbons in normally pressured chalk, filled microfossils. experience with respect to porosity-development 2 The cementation in the OJP may be promoted from overpressured chalk thus cannot be directly by Ca2+ diffusing from the basaltic basement, applied. whereas the chalk of Gorm and Tyra probably only has calcite dissolved at stylolites as a source of pore filling cement. CONCLUSIONS 3 The difference in burial history and temperature between central North Sea and OJP. In Stylolite formation is a major agent of calcite the studied North Sea fields, overpressure has dissolution causing calcite-cementation in chalk. lowered the rate of effective burial, and hence Stylolites have similar size in the chalk of the caused a low effective burial relative to tempera- Ontong Java Plateau and in the chalk of the Gorm ture. The silica of the North Sea chalk may ori- and Tyra fields of the central North Sea, where ginate from opal that has recrystallized to quartz effective stress is similar but temperature differs. at a stage where compaction was not completed These observations indicate that stylolite forma- and stylolites had not started forming. The sud- tion is controlled by effective stress. den cementation caused by the drop in silica in Stylolites tend to be localized at clay, probably the pore water of the deep-sea chalk may thus not because pressure dissolution is promoted where

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 203 calcite is squeezed into the electric double layer Mærsk Olie and Gas AS is acknowledged for data of clay with cation exchange capacity, so that ions and samples from the Gorm and Tyra fields. The may be conveyed away from the point of contact. following are thanked for discussions and help: In addition the presence of clay counteracts Finn Engstrøm, Morten L. Hjuler, Dieke Postma, precipitation of calcite. Susan Stipp, Henrik Tirsgaard, and Mads Wil- In the Ontong Java Plateau the stylolites lumsen. Technical assistance is acknowledged originiate in green colour bands containing smec- from: Iver Juel, Erik Leonardsen, Jørn Rønsbo, tite-chlorite and opal-A, which upon burial re- Kirsten Carlsen, Mimi Christensen, Hector Diaz, crystallize to smectite- and opal-A-bearing green Bente Frydenlund, Anna Syberg Hansen, Sinh flaser structures, where the first signs of pressure Nguyen,Tove Poulsen, and Mitzi Sørensen. Conx- dissolution are seen. Upon further burial they ita Taberner and an anonymous reviewer are recrystallize into smectite- and opal-CT-bearing thanked for constructive comments on a previous grey flaser structures and stylolites, which deeper version of the text. down are replaced by quartz- and illite-smectite- bearing stylolites. Whereas net dissolution of calcite is controlled REFERENCES by stress at stylolites, calcite precipitation in the Ontong Java Plateau is probably controlled by Bachman, R.T. (1984) Intratest porosity in foraminifera. J. Sed. temperature. The reason is that silica-diagenesis Petrol., 54, 257–262. Baker, P.A., Kastner, M., Byerlee, J.D. and Lockner, D.A. controls the amount of Si in the pore water, and (1980) Pressure dissolution and hydrothermal recrystalli- Ca–silicate complexes in the pore water may zation of carbonate sediments – an experimental study. Mar. cause apparent super-saturation with calcite. Geol., 38, 185–203. When opal is transformed to quartz due to Baker, P.A., Gieskes, J.M. and Elderfield, H. (1982) Diagenesis of carbonates in deep-sea sediments – evidence from Sr/Ca increasing temperature, the silica content in the 2+ pore water declines, and calcite precipitates as a ratios and interstitial dissolved Sr data. J. Sed. Petrol., 52, 71–82. cementation front. Bathurst, R.G.C. (1995) Burial diagenesis of limestones under In the Ontong Java Plateau, above the cemen- simple overburden. Stylolites, cementation and feedback. tation front, variations in primary texture have Bull. Soc. Geol. Fr., 166, 181–192. little influence on total porosity but, in the deep Berger, W.H. and Lind, I.L. (1997) Abundance of color bands in Neogene carbonate sediments on Ontong Java Plateau: a cemented interval, wackestone has lower porosity proxy for sedimentation rate. Mar. Geol., 144, 1–8. than mudstone due to preferential cementation of Berner, R.A. (1975) The role of magnesium in the crystal microfossils. growth of calcite and aragonite from sea water. Geochim. In the central North Sea Gorm and Tyra fields, Cosmochim. Acta, 39, 489–504. overpressuring is reflected in a relatively low Bishop, J.K.B. (1988) The barite-opal-organic carbon association in oceanic particulate matter. Nature, 332, 341–343. effective stress for a given depth and conse- Bjørkum, P.A. (1996) How important is pressure in causing quently a high ratio between temperature and dissolution of quartz in sandstones? J. Sed. Res., A66, 147– effective stress. This has probably caused silica 154. transformation to quartz before the onset of Bjørkum, P.A., Oelkers, E.H., Nadeau, P.H., Walderhaug, O. stylolite formation and consequent direct calcite and Murphy, W.M. (1998) Porosity prediction in quartzose cementation. The chalk is partly cemented as sandstones as a function of time, temperature, depth, stylolite frequency, and hydrocarbon saturation. AAPG Bull., reflected in cemented microfossils and a porous 82, 637–648. matrix, apparently with little pore-filling ce- Borre, M. (1997) Porosity variation in Cenozoic and Upper ment. Cretaceous Chalk from the Ontong Java Plateau – a discus- Burial stress and temperature thus play distinct sion of image analysis data. In: Research in Petroleum Technology (Ed. M.F. Middleton), pp. 1–15. Nordic Petro- roles in burial diagenesis and porosity-develop- leum Series 3, A˚ s Norway. ment of chalk. Burial stress controls physical Borre, M. (1998) Ultrasonic velocity of North Sea Chalk – compaction and pressure dissolution, whereas predicting saturate data from dry. In: Research in Petroleum temperature controls recrystallization and cemen- Technology (Ed. M.F. Middleton). pp. 71–98. Nordic Pet- tation. roleum Series, 4, A˚ s Norway. Borre, M. and Fabricius, I.L. (1998) Chemical and mechanical processes during burial diagenesis of chalk. An interpret- ation based on specific surface data of deep-sea sediments. ACKNOWLEDGEMENTS Sedimentology, 45, 755–770. Borre, M., Lind, I. and Mortensen, J. (1997) Specific surface as The Ocean Drilling Program is acknowledged for a measure of Burial Diagenesis of Chalk. Zbl. Geol. Pala¨ ont. Teil 1, 1995, 1071–1078. data and samples from the Ontong Java Plateau.

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 204 I. L. Fabricius and M. K. Borre

Cronan, D.S. (1974) Authigenic minerals in deep-sea sedi- Musgrave, R.J., Delaney, M.L., Stax, R. and Tarduno, J.A. ments. In: The Sea, 5 (Ed. E.D. Goldberg), pp. 491–525. John (1993) Magnetic diagenesis, organic input, interstitial water Wiley & Sons, New York. chemistry, and paleomagnetic record of the carbonate se- Delaney, M.L. and Linn, L.J. (1993) Interstitial water and bulk quence of the Ontong Java Plateau. Proc. ODP Sci. Results, calcite chemistry, Leg 130, and calcite recrystallization. 130, 527–546. Proc. ODP Sci. Results, 130, 561–572. Neugebauer, J. (1974) Some aspects of cementation in chalk. Delaney, M.L. and Shipboard Scientific Party (1991) Inor- In: Pelagic Sediments: On Land and Under the Sea (Eds ganic geochemistry summary. Proc. ODP Init. Rep., 130, K.J. Hsu¨ and H.C. Jenkyns), Int. Assoc. Sedimentol. Spec. 549–551. Publ., 1, 149–176. Doyle, M.C. and Conlin, J.M. (1990) The Tyra Field. In: North Oelkers, E.H., Bjørkum, P.A. and Murphy, W.M. (1992) The Sea Oil and Gas Reservoirs II (Eds A.T. Buller, E. Berg, O. mechanism of porosity reduction, stylolite development Hjelmeland, J. Kleppe, O. Torsæter and J.O. Aasen), pp. 47– and quartz cementation in North Sea sanstones. In: Water- 65. Graham & Trotman, London. Rock Interaction 2 (Eds Y.K. Kharaka and A.S. Maest), Dymond, J., Suess, E. and Lyle, M. (1992) Barium in deep-sea pp. 1183–1186. Balkema, Rotterdam. sediment: a geochemical proxy for paleoproductivity. Olsen, C. (2002) Texturel tolkning af Nordsøkalks elasticitet og Paleoceanography, 7, 163–181. porekollaps. MSc Thesis, Technical University of Denmark, Fabricius, I.L. (2003) How burial diagenesis of chalk sedi- Lyngby. ments controls sonic velocity and porosity. AAPG Bull., 87, Parkhurst, D.L. and Appelo, C.A.J. (1999) User’s Guide to 1755–1778. PHREEQC (version 2) – a Computer Program for Speciation, Heald, M.T. (1955) Stylolites in sandstone. J. Geol., 63, 101–114. Batch-Reaction, One-Dimensional Transport, and Inverse Hobert, L.A. and Wetzel, A. (1989) On the relationship be- Geochemical Calculations. U.S. Geol. Surv. Water Resour. tween silica and carbonate diagenesis in deep-sea sedi- Inv. Rep. 99-4259, 312 pp. ments. Geol. Rundsch., 78, 765–778. Røgen, B. (2002) North Sea chalk – textural, petrophysical, and Hurst, C. (1983) Petroleum geology of the Gorm field, Danish acoustic properties. PhD Thesis, Environment & Resources North Sea. Geol. Mijnbouw, 62, 157–168. DTU, Technical University of Denmark, Kgs. Lyngby. 106 Japsen, P. (1998) Regional velocity-depth anomalies, North pp. Sea Chalk: a record of overpressure and neogene uplift and Røgen, B. and Fabricius, I.L. (2002) Influence of clay and silica erosion. AAPG Bull., 82, 2031–2074. on permeability and capillary entry pressure of chalk re- Jensenius, J. (1987) High-temperature diagenesis in shallow servoirs in the North Sea. Petrol. Geosci., 8, 287–293. Chalk reservoir, Skjold oil field, Danish North Sea: evidence Røgen, B. and Fabricius, I.L. (2004) Diagenetic ripening of six from fluid inclusions and oxygen isotopes. AAPG Bull., 71, North Sea chalk fields. P334. In: Sharing the Earth. 66th 1378–1386. EAGE Conference and Exhibition, Paris, 7–20 June, 2004. Kastner, M. (1979) Authigenic silicates in deep-sea sediments: Extended abstracts and exhibitors’ catalogue. CD-ROM, formation and diagenesis. In: The Sea 7 (Ed. C. Emiliani), European Association of Geoscientists and Engineers, DB pp. 915–980. John Wiley & Sons, New York. Houten, NL, p. 4. Krissek, L.A. and Janecek, T.R. (1993) Eolian deposition on Røgen, B., Gommesen, L. and Fabricius, I.L. (2001) Grain size the Ontong Java Plateau since the Oligocene: unmixing the distributions of Chalk from image analysis of electron record of multiple dust sources. Proc. ODP Sci. Results, 130, micrographs. Comput. Geosci., 27, 1071–1080. 471–490. Renard, F. and Ortoleva, P. (1997) Water films at grain-grain Kroenke, L.W., Berger, W.H., Janecek, T.R. et al. (1991) Proc. contacts: Debye-Huckel, osmotic model of stress, salinity ODP Init. Rep., 130, 1240 pp. and mineralogy dependence. Geochim. Cosmochim. Acta, Lind, I.L. (1993a) Stylolites in chalk from Leg 130, Ontong Java 61, 1963–1970. Plateau. Proc. ODP Sci. Results, 130, 445–451. Renard, F., Brosse, E. and Gratier, J.P. (2000) The different Lind, I.L. (1993b) Loading experiments on carbonate ooze and processes involved in the mechanism of pressure solution chalk from Leg 130, Ontong Java Plateau. Proc. ODP Sci. in quartz-rich rocks and their interactions. In: Quartz Results, 130, 673–686. Cementation in Sandstones (Eds R.H. Worden and S. Mo- Lind, I.L., Janecek, T., Krissek, L., Prentice, M. and Stax, R. rad), Int. Assoc. Sedimentol. Spec. Publ., 29, 67–78. (1993) Color bands in Ontong Java Plateau carbonate oozes Safaricz, M. and Davison, I. (2005) Pressure solution in chalk. and chalks. Proc. ODP Sci. Results, 130, 453–470. AAPG Bull., 89, 383–401. Lind, I., Nykjær, O., Priisholm, S. and Springer, N. (1994) Schlanger, S.O. and Douglas, R.G. (1974) The pelagic ooze- Permeability of stylolite-bearing chalk. J. Petrol. Technol., chalk-limestone transition and its implications for marine 46, 986–993. stratigraphy. In: Pelagic Sediments: On Land and Under the Mahoney, J.J., Fitton, J.G., Wallace, P.J. et al. (2001) Proc. Sea (Eds K.J. Hsu¨ and H.C. Jenkyns). Int. Assoc. Sedimentol. ODP Init. Rep., 192, CD-ROM. Spec. Publ., 1, 117–148. Mazzullo, J., Meyer, A. and Kidd, R. (1988) New sediment Scholle, P.A. (1977) Chalk diagenesis and its relation to pet- classification scheme for the Ocean Drilling Program. In: roleum exploration: oil from chalks, a modern miracle? Handbook for Shipboard Sedimentologists (Eds J. Mazzullo AAPG Bull., 61, 982–1009. and A.G. Graham) Texas A&M University, Ocean Drilling Sheldon, H., Wheeler, J., Worden, R.H. and Cheadle, M.J. Program, Technical Note, 8, 45–67. (2003) An analysis of the roles of stress, temperature, and Moberly, R., Schlanger, S.O. et al. (1986) Init. Rep. DSDP, 89. pH in chemical compaction of sandstones. J. Sed. Res., 73, US Govt. Printing Office, Washington, DC. 678 pp. 64–71. Mortensen, J., Engstrøm, F. and Lind, I. (1998) The relation Sheppard, T.H. (2002) Stylolite development at sites of pri- among porosity, permeability, and specific surface of chalk mary and diagenetic fabric contrast within the Sutton Stone from the Gorm field, Danish North Sea. SPE Reservoir Eval. (Lower Lias), Ogmore-by-Sea, Glamorgan, UK. Proc. Geol. Eng., 1, 245–251. Assoc., 113, 87–109.

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205 Stylolites in Ontong Java Plateau chalk and North Sea chalk 205

Simpson, J. (1985) Stylolite-controlled layering in an homo- area: results from ODP Sites 680, 682, 685, and 688. Proc. geneous limestone: pseudo-bedding produced by burial ODP Sci. Results, 112, 491–503. diagenesis. Sedimentology, 32, 495–505. Weyl, P.K. (1959) Pressure solution and the force of crystal- Tanaka, M. and Takahashi, K. (2000) The identiﬁcation and lization – a phenomenological theory. J. Geophys. Res., 64, characterization of silicate complexes in calcium chloride 2001–2025. solution using fast atom bombardment mass spectrometry. Winterer, E.L. et al. (1971) Init. Rep. DSDP, 7. US Govt. Prin- Anal. Chim. Acta, 411, 109–119. ting Ofﬁce, Washington, DC. 930 pp. Tanaka, M. and Takahashi, K. (2002) Characterization of silicate monomer with sodium, calcium and strontium but not Manuscript received 12 October 2005; revision with lithium and magnesium ions by fast atom bombardment accepted 9 August 2006 mass spectrometry. J. Mass Spectrom., 37, 623–630. Von Breymann, M.T., Emeis, K.-Ch. and Camerlenghi, A. (1989) Geochemistry of sediments from the Peru upwelling

2006 The Authors. Journal compilation 2006 International Association of Sedimentologists, Sedimentology, 54, 183–205